基于电力大数据的变电设备故障诊断方法研究

Research on fault detection method of substation equipment based on power big data

针对变电设备故障具有随机性、诊断过程复杂等问题,基于人工蜂群算法、二叉树和支持向量机模型,提出一种基于电力大数据的变电设备故障诊断方法。首先通过人工蜂群算法对支持向量机的参数进行优化,然后针对变电设备故障类型多的特点构造了基于二叉树的多分类支持向量机,最后通过实例验证和分析该方法的有效性和分类性能。结果表明,所提出的故障诊断方法能够准确地实现变电设备故障分类。

Regarding the randomness of substation equipment faults and complex diagnosis process,based on Artificial Bee Colony algorithm,Binary tree and Support Vector Machine model,a fault diagnosis method for substation equipment based on power big data is proposed.Firstly,the parameters of SVM are optimized by Artificial Bee Colony algorithm,and then a Binary tree based multi-class SVM is constructed on the basis of the characteristics of multiple fault types of substation equipment.Finally,the effectiveness and classification performance of the method are verified and analyzed by examples.The results show that the proposed fault diagnosis method can accurately classify the fault types of substation equipment.